Summary
In this chapter, we introduced the concept of a parameterised quantum circuit as a generic QML model. PQCs can be trained and used as discriminative and generative QML models as well as optimisers. They can also be used to encode classical data samples into the corresponding quantum states.
We considered several popular data encoding methods. Arguably, the simplest and easiest to implement is the angle encoding algorithm – we shall use this approach in the next chapter. Other methods also have their strong points, although they tend to be either more demanding in terms of the hardware capabilities or better suited for some niche applications.
In the next chapter, we apply what we have learned so far to the task of building the quantum neural network trained as a classifier and compare its performance on the binary classification problem with standard classical machine learning models.
